About this Research Topic
Problematic substance use - e.g. use of alcohol, opioids, tobacco, stimulants - is among one of the leading contributors to the global burden of disease, with approximately 1 in 10 individuals developing a substance use disorder over their lifetime. Deaths due to substance use disorders have risen substantially over the last two decades, particularly in the United States, due to laissez-faire opioid prescribing practices and the availability of high-potency synthetic opioids. The economic impact of substance use disorder is estimated to be around US$740 billion dollars per year in the United States alone.
Currently, much published behavioral research on substance use disorders is based on survey data, with relatively little emphasis on the secondary use of existing data derived from social media or electronic health records. Given the large volume of health-related data available from social media platforms (e.g. Reddit, Twitter), online communities, and electronic health records, computational methods leveraged to collect and process these diverse data at scale can serve as a useful complement to traditional survey-based methods. Computational methods have demonstrated their value in the context of population-level substance use disorder and addiction research efforts, including such use cases as identifying changes in smoking trends from clinical text, identifying behavioral risk factors associated with potential opioid overdose from clinical text, exploring public perceptions of substances on Twitter, and investigating the non-medical use of psychostimulant drugs among college students using Twitter data. However, much is still left unknown.
We welcome and encourage contributions to this Research Topic that focus on the application of computational methods to diverse textual health data sources including but not limited to social media, online health communities, and electronic health records, with the broad goal of contributing to our current knowledge of substance use and addiction.
Topics of interest include:
· Identification of substance use status, duration of use, whether the use reported is problematic, and study personal and community-level risk factors;
· Characterize changes in users behaviors, attitudes regarding particular substances, and contextual factors and motivations associated with substance use;
· Longitudinal studies of social media users to better understand trajectories of substance use;
· Analyzing substance use stigma (and the effects of substance use stigma);
· Comparative effectiveness research on substance use and addiction treatments.
Currently, much published behavioral research on substance use disorders is based on survey data, with relatively little emphasis on the secondary use of existing data derived from social media or electronic health records. Given the large volume of health-related data available from social media platforms (e.g. Reddit, Twitter), online communities, and electronic health records, computational methods leveraged to collect and process these diverse data at scale can serve as a useful complement to traditional survey-based methods. Computational methods have demonstrated their value in the context of population-level substance use disorder and addiction research efforts, including such use cases as identifying changes in smoking trends from clinical text, identifying behavioral risk factors associated with potential opioid overdose from clinical text, exploring public perceptions of substances on Twitter, and investigating the non-medical use of psychostimulant drugs among college students using Twitter data. However, much is still left unknown.
We welcome and encourage contributions to this Research Topic that focus on the application of computational methods to diverse textual health data sources including but not limited to social media, online health communities, and electronic health records, with the broad goal of contributing to our current knowledge of substance use and addiction.
Topics of interest include:
· Identification of substance use status, duration of use, whether the use reported is problematic, and study personal and community-level risk factors;
· Characterize changes in users behaviors, attitudes regarding particular substances, and contextual factors and motivations associated with substance use;
· Longitudinal studies of social media users to better understand trajectories of substance use;
· Analyzing substance use stigma (and the effects of substance use stigma);
· Comparative effectiveness research on substance use and addiction treatments.
Keywords: natural language processing, substance use, addiction, social media, electronic health records
Important Note: All contributions to this Research Topic must be within the scope of the section and journal to which they are submitted, as defined in their mission statements. Frontiers reserves the right to guide an out-of-scope manuscript to a more suitable section or journal at any stage of peer review.
